The Effect of Acute Rises in Intraocular Pressure after Intravitreal Bevacizumab Injection on the Peripapillary Retinal Nerve Fiber Layer Thickness and the Role of Anterior Chamber Paracentesis.

PURPOSE: To evaluate acute changes in intraocular pressure (IOP) and their short-term effects on the peripapillary retinal nerve fiber layer (RNFL) thickness after intravitreal bevacizumab (IVB) injection.
METHODS: Fifty-eight eyes of 37 patients with treatment-naïve diabetic macular edema or exudative age-related macular degeneration were included in the study. Patients were divided into two groups, and the participants of each group received 3 monthly injections of IVB. IOP was measured right before the injection, immediately after the injection, and 5 min and 20 min after each injection. Peripapillary, RNFL thickness was measured before the injection and 1 month after the third injection. In the second group, anterior chamber (AC) paracentesis was performed before IVB injection.
RESULTS: IOP values after injection in all sessions were significantly higher in the first group (P < 0.001). The peripapillary RNFL thickness changes 1 month after the third injection was not statistically significantly different in each group (P = 0.816 and 0.773 for the first and second groups, respectively).
CONCLUSION: AC paracentesis is an effective modality to reduce the acute rise in IOP. The effect of acute IOP elevation on the peripapillary RNFL thickness was not statistically significant. Copyright:

INTRODUCTION

Diabetic retinopathy and age-related macular degeneration are among the most prevalent causes of blindness in adults. The standard treatment of diabetic macular edema (DME) and neovascular age-related macular degeneration (nAMD) is intravitreal anti-vascular endothelial growth factor (VEGF) injection.1 Due to the chronic nature of DME and nAMD, the most standard treatment regimens recommend multiple intravitreal anti-VEGF injections at least during the 1st year of treatment. One of the concerns in this regard is rising intraocular pressure (IOP) with the subsequent risk of glaucomatous optic nerve head (ONH) damage. These issues have been evaluated in several studies, but the results have been inconclusive, and the significance of the matter remains to be elucidated. Several hypotheses have been proposed to explain the acute IOP spike after intravitreal injections, including anterior chamber (AC) angle narrowing, trabecular meshwork contracture (which results in reduced outflow facility), trabecular clogging with large molecules and proteins, decreased intraocular nitrite oxide, and intraocular hydrostatic pressure elevation secondary to volume excess.123 To avoid acute IOP increases, authors have made several recommendations, including using AC paracentesis (the most effective method), pretreatment with anti-glaucoma medications, and ocular decompression (with a cotton-tipped applicator before intravitreal injection).14 However, the exact role of the AC paracentesis in preventing glaucomatous damages following intravitreal injection is not clear. Furthermore, it is an invasive procedure that is associated with complications such as traumatic cataract and endophthalmitis.45 Therefore, there is no consensus on this intervention for all patients who have undergone intravitreal anti-VEGF injection. In this study, we aimed to evaluate the effect of AC paracentesis on the IOP rise and peripapillary short-term retinal nerve fiber layer (RNFL) thickness changes following intravitreal injection to add more evidence to the literature.

METHODS

All procedures performed in the studies involving human participants followed the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Furthermore, all procedures were approved by the institutional review board committee of Tehran University of Medical Sciences (Ethical code: IR.TUMS.FARABIH.REC.1396.4334). Informed consent was obtained from all participants.

In this study, patients with treatment-naïve DME without proliferative diabetic retinopathy (PDR) and treatment-naïve nAMD were included in the study. Exclusion criteria were defined as a history of intravitreal injection, laser retinal photocoagulation, PDR, glaucoma disease, and severe media opacity. Patients were randomly assigned to Group A or Group B. Group A patients received only 1.25-mg intravitreal bevacizumab (IVB) injection. For patients in Group B, the intravitreal injection was preceded by AC paracentesis. All patients received 3 monthly injections. In each session, IOP was measured with Tonopen (Reichert, Depew, NY) with the patient in the supine position (i) right before injection, (ii) immediately after the injection, (iii) 5 min after the injection, and (iv) 20 min after the injection. Ocu-Film covers were sterilized with gas plasma technique and replaced for each IOP measurement.

For all the participants, the peripapillary RNFL thickness was measured using spectral-domain optical coherence tomography (SD-OCT) (Spectralis SD-OCT, Heidelberg Engineering, Germany) at the baseline and month 3 (1 month after the third IVB injection) with follow-up software available in the device. The peripapillary RNFL (particularly the temporal region may be affected by the wax and wane nature of macular edema) and on the other hand glaucoma has a propensity to damage inferotemporal and supratemporal more than other sectors. Therefore, we recorded the data of peripapillary sectors separately for further analysis.

All statistical analyses were performed by Stata (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX, USA: StataCorp LLC.). To compare the changes in RNFL values within the groups, we used a paired t-test with estimates of fixed effects. To compare the IOP values within the groups in each session, we used a repeated measure analysis with estimated marginal means. Comparison of the IOP between the two groups, adjusted for the baseline values, through the sessions was performed through a multilevel linear regression analysis with 4 levels (single measurements, each time, each session, each subject, and correspondingly form level 1 to level 4). Furthermore, the comparison of the baseline values was performed in this analysis. P < 0.05 were considered statistically significant.

RESULTS

Fifty-eight eyes of 37 patients (30 eyes from Group A and 28 eyes from Group B) were included in the study. The mean age of the patients was 62.9 ± 10 years, and the median age was 64 years (range, 34–82 year). Sixteen (43.3%) patients were male, and 21 (56.8%) were female. Of the 58 eyes, 10 (17.2%) had nAMD, and 48 (82.8%) had DME. The baseline means pre-IOP values were 15 ± 3 and 16 ± 3 mmHg in Group A and Group B, respectively, during the first session. This difference was not statistically significant (P = 0.233). Mean immediate post-IOP, 5 min IOP, and 20 min IOP values were 37 ± 8, 31 ± 9, and 24 ± 7 mmHg, respectively, for Group A and 19 ± 5, 14 ± 3, and 14 ± 3 mmHg for Group B. The values of all postinjection measurements were significantly lower in Group B (P < 0.001 for all measurements).

A similar finding was observed for the second and third injections [Figure 1]. In Group A, the 20 min IOP values remained significantly higher than pre-IOP measurements for all sessions (mean differences from the baseline for the first, second, and third sessions were 8.667, 10.700, and 9.567, respectively, (P < 0.001 for all sessions). In contrast, 20 min IOP was not significantly different from pre-IOP measurements for Group B for the first (P = 0.060) and third (P = 0.162) sessions. However, the 20 min IOP was significantly lower than that of the pre-IOP measurement for the second session (P = 0.041).

Figure 1

Mean intraocular pressure in each group. Pre: Preinjection, Post: Postinjection, 5 min: 5 minutes after injection, 20 min: 20 minutes after injection

The mean global peripapillary RNFL thicknesses before and after treatment in Group A were 110 ± 19 and 111 ± 18 μ, respectively. The changes were not statistically significant (P = 0.816). A similar finding was evident in Group B (115 ± 19 μ, P = 0.773). Furthermore, the peripapillary RNFL thicknesses in the temporal superior, temporal inferior, nasal superior, and nasal inferior sectors at month 3 were not significantly different in either group when compared to the baseline values [Table 1]. Likewise, changes in the mean peripapillary RNFL thickness from the baseline values were not statistically significant between the two groups (P: 0.857, 0.718, 0.332, 0.652, and 0.142 for global, temporal superior, nasal superior, temporal inferior, and nasal inferior thicknesses, respectively).

Table 1

Mean peripapillary retinal nerve fiber layer thickness before (-pre) and 1 month after the third (-post) intravitreal bevacizumab injection

	Group A			Group B
	Mean thickness*	SD	P**	Mean thickness*	SD	P**
Global-pre	110	19	0.816	115	19	0.773
Global-post	111	18		115	19
TS-pre	145	24	0.198	141	25	0.446
TS-post	141	24		140	26
NS-pre	124	27	0.604	121	25	0.842
NS-post	141	24		140	26
TI-pre	149	37	0.126	155	35	0.692
TI-post	144	33		154	32
NI-pre	125	37	0.849	137	24	0.295
NI-post	124	24		141	35

*Micrometre, **Paired sample t-test. TS: Temporal superior, NS: Nasal superior, TI: Temporal inferior, NI: Nasal inferior, SD: Standard deviation

DISCUSSION

The intravitreal injection of anti-VEGFs (most of which require multiple injections) has become the choice of treatment in various retinal diseases. One of the concerns in this regard is acute and sustained IOP increases after the injection of anti-VEGFs. Some authors believe that multiple IOP spikes or sustained elevated IOP could induce glaucomatous damage, and so they have proposed several interventions to prevent IOP spikes after intravitreal injection.67 Among them, AC paracentesis is the most effective way to reduce postinjection IOP spikes. However, there is not enough evidence regarding the efficacy and safety of such methods, nor is there any consistency among studies on this treatment method.4

In this study, Group B showed significantly lower values of IOP after IVB when compared to Group A. This finding was evident in all measurements (0 min, 5 min, and 20 min) across all three sessions. Figure 1 depicts the time course of the acute IOP rise after injection. In Group A, IOP increased acutely and then followed a downward slope, but after 20 min, it remained significantly higher than that of the baseline. A similar pattern has been described by other studies.1 In a study by Soheilian et al., the mean IOP after IVB injection remained significantly higher at 30 min in a group without AC paracentesis). Meanwhile, in a group that received AC paracentesis, IOP decreased at a level that remained statistically significant for up to 3 months.8 In addition, they showed that mean RNFL thickness changes over 3 months were statistically significant between two groups (−2 vs. 0 μ, P < 0.001), and they suggested that AC paracentesis prevents peripapillary RNFL loss.8 This finding is in contrast to ours, as we found that the peripapillary RNFL thickness profile does not change significantly in all sectors. This may be because their participants were older than ours (66 year vs. 62 year) and had thinner baseline RNFL (although it was measured with a different instrument). These factors would make their patients more sensitive to IOP fluctuations. However, most studies have shown that RNFL does not change significantly after the intravitreal injection of anti-VEGF drugs,9101112131415 while other reports showed significant RNFL loss related to multiple intravitreal injections.816 This discrepancy across studies may be due to different baseline characteristics of patients, follow-up times, or the number of injections.

On the other hand, the adverse effects of the increased IOP are not limited to RNFL loss and glaucoma damages. Abdolrahimzadeh et al. observed macular ganglion cell layer loss after the intravitreal injection of anti-VEGF agents for nAMD. However, the same was not seen in the study by Zucchiatti et al.1417 Further, Wen et al. observed acute reductions in ONH perfusion parameters after intravitreal injections,18 and Mitsch et al. reported impaired vascular autoregulation and oxygen distribution in patients with DME after intravitreal injection of the anti-VEGF agent.19 Hence, all of them could affect patients' visual function. The short-term stability of RNFL does not obviate the serious consequences of IOP peaks, especially for most retinal conditions that require multiple injections.

In our study, vision-threatening complications (traumatic cataract, vitreous hemorrhage, retinal detachment, and endophthalmitis,) due to intravitreal injection and AC paracentesis were not seen. It was comparable with prior studies which have demonstrated the safety of AC paracentesis before intravitreal injections without sight-threatening complications.52021

A significant strength of our study is related to its design (randomized controlled) and the fact that we measured IOP in each of three sessions, thus increasing the validity of the study. In addition, peripapillary RNFL profiles were analyzed sector by sector, which allowed us to detect any changes in RNFL that could be missed by total measurements.

However, the study has some limitations as well due to the small sample size, lack of power analysis, short-term evaluation of IOP, short follow-up time, and the RNFL thickness (particular temporal sector), all of which might be affected by macular edema fluctuations. A larger randomized controlled study with a longer follow-up time is required to assess the efficacy and safety of AC paracentesis under these conditions.

In conclusion, there are no statistically significant changes in the peripapillary RNFL thickness after 3 monthly IVB injections. AC paracentesis seems not needed routinely before IVB injection since the short-term effects of IOP spikes on the peripapillary RNFL are not significant.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.